Bayesian Analysis
library(rstan)## Loading required package: ggplot2## Loading required package: StanHeaders## rstan (Version 2.18.2, GitRev: 2e1f913d3ca3)## For execution on a local, multicore CPU with excess RAM we recommend calling
## options(mc.cores = parallel::detectCores()).
## To avoid recompilation of unchanged Stan programs, we recommend calling
## rstan_options(auto_write = TRUE)library(ggmcmc)## Loading required package: dplyr##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union## Loading required package: tidyr##
## Attaching package: 'tidyr'## The following object is masked from 'package:rstan':
##
## extractlibrary(bayesplot)## This is bayesplot version 1.6.0## - Online documentation and vignettes at mc-stan.org/bayesplot## - bayesplot theme set to bayesplot::theme_default()## * Does _not_ affect other ggplot2 plots## * See ?bayesplot_theme_set for details on theme settinglibrary(bridgesampling)load("data_for_stan.RData")
list_for_stan <- list(nY = nrow(data_for_stan), nS = max(data_for_stan$participant), Subj = data_for_stan$participant, size_diff = data_for_stan$size, num_of_trials = data_for_stan$number_of_fixational_trials, Y = data_for_stan$answer.keys)Оценка связи типа иллюзии и разницы между стимулами
fit_1 <- stan("bayes_model_1.stan", data = list_for_stan, iter = 6000, warmup = 2000)##
## SAMPLING FOR MODEL 'bayes_model_1' NOW (CHAIN 1).
## Chain 1:
## Chain 1: Gradient evaluation took 0.000102 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 1.02 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 1: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 1: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 1: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 1: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 1: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 1: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 1: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 1: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 1: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 1: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 1: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 1:
## Chain 1: Elapsed Time: 1.01642 seconds (Warm-up)
## Chain 1: 2.49926 seconds (Sampling)
## Chain 1: 3.51568 seconds (Total)
## Chain 1:
##
## SAMPLING FOR MODEL 'bayes_model_1' NOW (CHAIN 2).
## Chain 2:
## Chain 2: Gradient evaluation took 3.9e-05 seconds
## Chain 2: 1000 transitions using 10 leapfrog steps per transition would take 0.39 seconds.
## Chain 2: Adjust your expectations accordingly!
## Chain 2:
## Chain 2:
## Chain 2: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 2: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 2: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 2: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 2: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 2: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 2: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 2: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 2: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 2: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 2: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 2: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 2:
## Chain 2: Elapsed Time: 0.965543 seconds (Warm-up)
## Chain 2: 2.53149 seconds (Sampling)
## Chain 2: 3.49704 seconds (Total)
## Chain 2:
##
## SAMPLING FOR MODEL 'bayes_model_1' NOW (CHAIN 3).
## Chain 3:
## Chain 3: Gradient evaluation took 3.8e-05 seconds
## Chain 3: 1000 transitions using 10 leapfrog steps per transition would take 0.38 seconds.
## Chain 3: Adjust your expectations accordingly!
## Chain 3:
## Chain 3:
## Chain 3: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 3: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 3: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 3: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 3: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 3: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 3: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 3: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 3: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 3: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 3: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 3: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 3:
## Chain 3: Elapsed Time: 1.03701 seconds (Warm-up)
## Chain 3: 2.29807 seconds (Sampling)
## Chain 3: 3.33508 seconds (Total)
## Chain 3:
##
## SAMPLING FOR MODEL 'bayes_model_1' NOW (CHAIN 4).
## Chain 4:
## Chain 4: Gradient evaluation took 4.2e-05 seconds
## Chain 4: 1000 transitions using 10 leapfrog steps per transition would take 0.42 seconds.
## Chain 4: Adjust your expectations accordingly!
## Chain 4:
## Chain 4:
## Chain 4: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 4: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 4: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 4: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 4: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 4: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 4: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 4: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 4: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 4: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 4: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 4: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 4:
## Chain 4: Elapsed Time: 1.06415 seconds (Warm-up)
## Chain 4: 2.57186 seconds (Sampling)
## Chain 4: 3.636 seconds (Total)
## Chain 4:print(fit_1)## Inference for Stan model: bayes_model_1.
## 4 chains, each with iter=6000; warmup=2000; thin=1;
## post-warmup draws per chain=4000, total post-warmup draws=16000.
##
## mean se_mean sd 2.5% 25% 50% 75% 97.5%
## intercept 1.06 0.01 0.43 0.24 0.77 1.05 1.35 1.93
## beta_mu -0.94 0.00 0.32 -1.60 -1.14 -0.92 -0.72 -0.34
## beta_sd 1.12 0.00 0.27 0.70 0.93 1.09 1.28 1.76
## beta[1] -0.48 0.00 0.34 -1.15 -0.71 -0.48 -0.25 0.19
## beta[2] -0.06 0.00 0.48 -0.97 -0.39 -0.08 0.24 0.95
## beta[3] -1.75 0.01 0.68 -3.32 -2.14 -1.67 -1.28 -0.67
## beta[4] -2.40 0.01 0.85 -4.34 -2.89 -2.29 -1.78 -1.06
## beta[5] -1.33 0.01 0.52 -2.47 -1.65 -1.28 -0.96 -0.42
## beta[6] 0.39 0.01 0.58 -0.56 0.00 0.34 0.72 1.70
## beta[7] -0.01 0.00 0.38 -0.74 -0.27 -0.03 0.23 0.76
## beta[8] -0.04 0.00 0.36 -0.70 -0.28 -0.05 0.20 0.70
## beta[9] -0.38 0.00 0.40 -1.16 -0.65 -0.38 -0.11 0.43
## beta[10] -2.13 0.01 0.85 -4.09 -2.60 -2.01 -1.53 -0.81
## beta[11] -1.49 0.01 0.46 -2.49 -1.77 -1.45 -1.18 -0.69
## beta[12] -1.46 0.01 0.50 -2.52 -1.76 -1.42 -1.11 -0.59
## beta[13] -1.69 0.01 0.90 -3.71 -2.19 -1.59 -1.06 -0.26
## beta[14] -1.09 0.00 0.40 -1.94 -1.34 -1.06 -0.81 -0.38
## beta[15] -2.33 0.01 0.83 -4.27 -2.79 -2.22 -1.75 -1.05
## beta[16] -0.88 0.00 0.36 -1.62 -1.10 -0.86 -0.63 -0.22
## beta[17] 0.09 0.00 0.36 -0.58 -0.16 0.07 0.32 0.84
## beta[18] -1.33 0.01 0.74 -2.99 -1.77 -1.26 -0.81 -0.06
## beta[19] -2.46 0.01 0.82 -4.37 -2.91 -2.35 -1.89 -1.20
## beta[20] -0.16 0.00 0.34 -0.82 -0.39 -0.17 0.06 0.54
## beta[21] -1.63 0.01 0.88 -3.68 -2.12 -1.52 -1.02 -0.23
## beta[22] -1.30 0.00 0.43 -2.26 -1.57 -1.27 -1.00 -0.54
## beta[23] 0.25 0.00 0.35 -0.38 0.00 0.23 0.47 1.00
## beta[24] 0.58 0.00 0.47 -0.22 0.25 0.54 0.86 1.64
## beta[25] -0.37 0.00 0.47 -1.30 -0.68 -0.37 -0.06 0.55
## beta[26] -2.15 0.01 0.87 -4.14 -2.63 -2.04 -1.54 -0.76
## lp__ -124.29 0.08 4.85 -134.92 -127.28 -123.90 -120.90 -115.92
## n_eff Rhat
## intercept 4215 1
## beta_mu 5378 1
## beta_sd 4755 1
## beta[1] 8271 1
## beta[2] 12277 1
## beta[3] 8287 1
## beta[4] 7443 1
## beta[5] 9931 1
## beta[6] 11904 1
## beta[7] 9502 1
## beta[8] 9462 1
## beta[9] 11787 1
## beta[10] 7840 1
## beta[11] 8144 1
## beta[12] 8523 1
## beta[13] 10069 1
## beta[14] 9490 1
## beta[15] 8002 1
## beta[16] 9101 1
## beta[17] 9959 1
## beta[18] 11723 1
## beta[19] 7063 1
## beta[20] 10318 1
## beta[21] 10001 1
## beta[22] 8502 1
## beta[23] 9879 1
## beta[24] 10404 1
## beta[25] 12476 1
## beta[26] 7888 1
## lp__ 3422 1
##
## Samples were drawn using NUTS(diag_e) at Fri Jan 4 16:42:49 2019.
## For each parameter, n_eff is a crude measure of effective sample size,
## and Rhat is the potential scale reduction factor on split chains (at
## convergence, Rhat=1).posterior_1 <- as.array(fit_1)
mcmc_areas(posterior_1, regex_pars = c("beta"), prob = 0.8, prob_outer = 0.95, point_est = "mean")
Диагностика
posterior_1b <- ggs(fit_1)
ggmcmc(D = posterior_1b, file = NULL, plot = 'ggs_traceplot') ## Plotting traceplots
## Time taken to generate the report: 16 seconds.ggmcmc(D = posterior_1b, file = NULL, plot = 'ggs_compare_partial') ## Plotting comparison of partial and full chain
## Time taken to generate the report: 19 seconds.ggmcmc(D = posterior_1b, file = NULL, plot = 'ggs_autocorrelation') ## Plotting autocorrelation plots
## Time taken to generate the report: 15 seconds.Модель с индивидуальными смещениями
fit_2 <- stan("bayes_model_2.stan", data = list_for_stan, iter = 6000, warmup = 2000)##
## SAMPLING FOR MODEL 'bayes_model_2' NOW (CHAIN 1).
## Chain 1: Rejecting initial value:
## Chain 1: Error evaluating the log probability at the initial value.
## Chain 1: Exception: beta_lpdf: Random variable is -1.45713, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 1:
## Chain 1: Gradient evaluation took 0.000139 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 1.39 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 1: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 1: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 1: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 1: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 1: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 1: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 1: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 1: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 1: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 1: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 1: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 1:
## Chain 1: Elapsed Time: 1.04328 seconds (Warm-up)
## Chain 1: 1.8008 seconds (Sampling)
## Chain 1: 2.84408 seconds (Total)
## Chain 1:
##
## SAMPLING FOR MODEL 'bayes_model_2' NOW (CHAIN 2).
## Chain 2:
## Chain 2: Gradient evaluation took 4.3e-05 seconds
## Chain 2: 1000 transitions using 10 leapfrog steps per transition would take 0.43 seconds.
## Chain 2: Adjust your expectations accordingly!
## Chain 2:
## Chain 2:
## Chain 2: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 2: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 2: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 2: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 2: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 2: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 2: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 2: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 2: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 2: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 2: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 2: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 2:
## Chain 2: Elapsed Time: 1.15235 seconds (Warm-up)
## Chain 2: 1.67036 seconds (Sampling)
## Chain 2: 2.82271 seconds (Total)
## Chain 2:
##
## SAMPLING FOR MODEL 'bayes_model_2' NOW (CHAIN 3).
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is 1.19705, but must be less than or equal to 1 (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is -0.336032, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is 1.52124, but must be less than or equal to 1 (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is -0.0588813, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is -0.963093, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is 1.5279, but must be less than or equal to 1 (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 3:
## Chain 3: Gradient evaluation took 4e-05 seconds
## Chain 3: 1000 transitions using 10 leapfrog steps per transition would take 0.4 seconds.
## Chain 3: Adjust your expectations accordingly!
## Chain 3:
## Chain 3:
## Chain 3: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 3: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 3: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 3: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 3: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 3: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 3: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 3: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 3: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 3: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 3: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 3: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 3:
## Chain 3: Elapsed Time: 1.26678 seconds (Warm-up)
## Chain 3: 2.88996 seconds (Sampling)
## Chain 3: 4.15674 seconds (Total)
## Chain 3:
##
## SAMPLING FOR MODEL 'bayes_model_2' NOW (CHAIN 4).
## Chain 4: Rejecting initial value:
## Chain 4: Error evaluating the log probability at the initial value.
## Chain 4: Exception: beta_lpdf: Random variable is -0.264047, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 4: Rejecting initial value:
## Chain 4: Error evaluating the log probability at the initial value.
## Chain 4: Exception: beta_lpdf: Random variable is 1.2442, but must be less than or equal to 1 (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 4: Rejecting initial value:
## Chain 4: Error evaluating the log probability at the initial value.
## Chain 4: Exception: beta_lpdf: Random variable is -1.0056, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 4: Rejecting initial value:
## Chain 4: Error evaluating the log probability at the initial value.
## Chain 4: Exception: beta_lpdf: Random variable is -0.0367961, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 4: Rejecting initial value:
## Chain 4: Error evaluating the log probability at the initial value.
## Chain 4: Exception: beta_lpdf: Random variable is -1.9334, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 4: Rejecting initial value:
## Chain 4: Error evaluating the log probability at the initial value.
## Chain 4: Exception: beta_lpdf: Random variable is -1.09291, but must be >= 0! (in 'model47ca61cf6d02_bayes_model_2' at line 20)
##
## Chain 4:
## Chain 4: Gradient evaluation took 6.2e-05 seconds
## Chain 4: 1000 transitions using 10 leapfrog steps per transition would take 0.62 seconds.
## Chain 4: Adjust your expectations accordingly!
## Chain 4:
## Chain 4:
## Chain 4: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 4: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 4: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 4: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 4: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 4: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 4: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 4: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 4: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 4: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 4: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 4: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 4:
## Chain 4: Elapsed Time: 1.21527 seconds (Warm-up)
## Chain 4: 1.7629 seconds (Sampling)
## Chain 4: 2.97817 seconds (Total)
## Chain 4:## Warning: There were 9082 divergent transitions after warmup. Increasing adapt_delta above 0.8 may help. See
## http://mc-stan.org/misc/warnings.html#divergent-transitions-after-warmup## Warning: Examine the pairs() plot to diagnose sampling problemsprint(fit_2)## Inference for Stan model: bayes_model_2.
## 4 chains, each with iter=6000; warmup=2000; thin=1;
## post-warmup draws per chain=4000, total post-warmup draws=16000.
##
## mean se_mean sd 2.5% 25% 50% 75% 97.5%
## intercept_mu 0.47 0.01 0.28 0.03 0.24 0.46 0.70 0.97
## intercept_sd 2.05 0.01 0.46 1.32 1.72 1.99 2.31 3.10
## intercept[1] 1.52 0.01 0.65 0.26 1.08 1.52 1.96 2.81
## intercept[2] 2.54 0.02 0.99 0.79 1.86 2.48 3.15 4.67
## intercept[3] -0.67 0.01 0.98 -2.72 -1.29 -0.61 0.02 1.07
## intercept[4] -2.20 0.02 1.34 -5.23 -2.95 -2.02 -1.26 -0.07
## intercept[5] -0.18 0.01 0.82 -1.91 -0.70 -0.15 0.36 1.38
## intercept[6] 2.48 0.02 1.02 0.64 1.79 2.41 3.11 4.64
## intercept[7] 2.23 0.01 0.71 0.92 1.74 2.21 2.68 3.75
## intercept[8] 2.11 0.01 0.76 0.67 1.58 2.09 2.61 3.64
## intercept[9] 1.94 0.01 0.87 0.32 1.35 1.90 2.50 3.72
## intercept[10] -1.79 0.02 1.41 -5.04 -2.57 -1.64 -0.81 0.52
## intercept[11] -0.82 0.01 0.78 -2.48 -1.31 -0.78 -0.29 0.59
## intercept[12] 0.06 0.01 0.65 -1.26 -0.37 0.08 0.51 1.31
## intercept[13] -0.91 0.02 1.59 -4.46 -1.88 -0.78 0.20 1.84
## intercept[14] -0.29 0.01 0.83 -2.07 -0.81 -0.24 0.26 1.25
## intercept[15] -2.15 0.02 1.33 -5.23 -2.88 -1.98 -1.22 0.01
## intercept[16] 0.23 0.01 0.75 -1.29 -0.26 0.25 0.75 1.66
## intercept[17] 2.62 0.02 0.83 1.09 2.05 2.59 3.14 4.34
## intercept[18] 0.32 0.02 1.18 -2.10 -0.44 0.35 1.12 2.53
## intercept[19] -2.42 0.02 1.30 -5.52 -3.12 -2.24 -1.50 -0.34
## intercept[20] 1.61 0.01 0.84 -0.01 1.04 1.61 2.19 3.28
## intercept[21] -0.76 0.03 1.68 -4.59 -1.72 -0.62 0.39 2.11
## intercept[22] 0.09 0.01 0.65 -1.19 -0.34 0.11 0.53 1.34
## intercept[23] 3.01 0.02 0.82 1.52 2.45 2.96 3.55 4.77
## intercept[24] 2.90 0.01 0.81 1.41 2.34 2.86 3.41 4.61
## intercept[25] 1.41 0.01 0.94 -0.40 0.79 1.39 2.02 3.31
## intercept[26] -1.78 0.03 1.47 -5.12 -2.57 -1.60 -0.78 0.55
## beta -0.57 0.00 0.17 -0.90 -0.69 -0.58 -0.46 -0.24
## lp__ -142.22 0.09 4.50 -151.99 -145.01 -141.90 -139.01 -134.43
## n_eff Rhat
## intercept_mu 2148 1
## intercept_sd 2793 1
## intercept[1] 3557 1
## intercept[2] 3477 1
## intercept[3] 4954 1
## intercept[4] 4627 1
## intercept[5] 4829 1
## intercept[6] 4304 1
## intercept[7] 3547 1
## intercept[8] 2925 1
## intercept[9] 3974 1
## intercept[10] 4124 1
## intercept[11] 3553 1
## intercept[12] 3857 1
## intercept[13] 5360 1
## intercept[14] 3673 1
## intercept[15] 4555 1
## intercept[16] 3396 1
## intercept[17] 3053 1
## intercept[18] 6034 1
## intercept[19] 4446 1
## intercept[20] 3712 1
## intercept[21] 4482 1
## intercept[22] 3585 1
## intercept[23] 2947 1
## intercept[24] 3091 1
## intercept[25] 4606 1
## intercept[26] 3256 1
## beta 1500 1
## lp__ 2751 1
##
## Samples were drawn using NUTS(diag_e) at Fri Jan 4 16:47:12 2019.
## For each parameter, n_eff is a crude measure of effective sample size,
## and Rhat is the potential scale reduction factor on split chains (at
## convergence, Rhat=1).posterior_2 <- as.array(fit_2)
mcmc_areas(posterior_2, regex_pars = c("intercept"), prob = 0.8, prob_outer = 0.95, point_est = "mean")
Диагностика
posterior_2b <- ggs(fit_2)
ggmcmc(D = posterior_2b, file = NULL, plot = 'ggs_traceplot') ## Plotting traceplots
## Time taken to generate the report: 14 seconds.ggmcmc(D = posterior_2b, file = NULL, plot = 'ggs_compare_partial') ## Plotting comparison of partial and full chain
## Time taken to generate the report: 18 seconds.ggmcmc(D = posterior_2b, file = NULL, plot = 'ggs_autocorrelation') ## Plotting autocorrelation plots
## Time taken to generate the report: 15 seconds.Модель с индивидуальными смещениями и индивидуальными размерами эффекта
fit_3 <- stan("bayes_model_3.stan", data = list_for_stan, iter = 6000, warmup = 2000)##
## SAMPLING FOR MODEL 'bayes_model_3' NOW (CHAIN 1).
## Chain 1:
## Chain 1: Gradient evaluation took 5.6e-05 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 0.56 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 1: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 1: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 1: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 1: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 1: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 1: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 1: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 1: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 1: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 1: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 1: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 1:
## Chain 1: Elapsed Time: 1.91152 seconds (Warm-up)
## Chain 1: 2.86213 seconds (Sampling)
## Chain 1: 4.77364 seconds (Total)
## Chain 1:
##
## SAMPLING FOR MODEL 'bayes_model_3' NOW (CHAIN 2).
## Chain 2: Rejecting initial value:
## Chain 2: Error evaluating the log probability at the initial value.
## Chain 2: Exception: beta_lpdf: Random variable is 1.16891, but must be less than or equal to 1 (in 'model47ca2e7c7cdd_bayes_model_3' at line 21)
##
## Chain 2: Rejecting initial value:
## Chain 2: Error evaluating the log probability at the initial value.
## Chain 2: Exception: beta_lpdf: Random variable is 1.61189, but must be less than or equal to 1 (in 'model47ca2e7c7cdd_bayes_model_3' at line 21)
##
## Chain 2: Rejecting initial value:
## Chain 2: Error evaluating the log probability at the initial value.
## Chain 2: Exception: beta_lpdf: Random variable is -0.947509, but must be >= 0! (in 'model47ca2e7c7cdd_bayes_model_3' at line 21)
##
## Chain 2: Rejecting initial value:
## Chain 2: Error evaluating the log probability at the initial value.
## Chain 2: Exception: beta_lpdf: Random variable is -1.26298, but must be >= 0! (in 'model47ca2e7c7cdd_bayes_model_3' at line 21)
##
## Chain 2:
## Chain 2: Gradient evaluation took 4.6e-05 seconds
## Chain 2: 1000 transitions using 10 leapfrog steps per transition would take 0.46 seconds.
## Chain 2: Adjust your expectations accordingly!
## Chain 2:
## Chain 2:
## Chain 2: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 2: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 2: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 2: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 2: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 2: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 2: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 2: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 2: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 2: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 2: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 2: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 2:
## Chain 2: Elapsed Time: 1.86638 seconds (Warm-up)
## Chain 2: 3.48361 seconds (Sampling)
## Chain 2: 5.34999 seconds (Total)
## Chain 2:
##
## SAMPLING FOR MODEL 'bayes_model_3' NOW (CHAIN 3).
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is 1.24996, but must be less than or equal to 1 (in 'model47ca2e7c7cdd_bayes_model_3' at line 21)
##
## Chain 3: Rejecting initial value:
## Chain 3: Error evaluating the log probability at the initial value.
## Chain 3: Exception: beta_lpdf: Random variable is 1.87894, but must be less than or equal to 1 (in 'model47ca2e7c7cdd_bayes_model_3' at line 21)
##
## Chain 3:
## Chain 3: Gradient evaluation took 4.2e-05 seconds
## Chain 3: 1000 transitions using 10 leapfrog steps per transition would take 0.42 seconds.
## Chain 3: Adjust your expectations accordingly!
## Chain 3:
## Chain 3:
## Chain 3: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 3: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 3: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 3: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 3: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 3: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 3: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 3: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 3: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 3: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 3: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 3: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 3:
## Chain 3: Elapsed Time: 2.20559 seconds (Warm-up)
## Chain 3: 3.05649 seconds (Sampling)
## Chain 3: 5.26208 seconds (Total)
## Chain 3:
##
## SAMPLING FOR MODEL 'bayes_model_3' NOW (CHAIN 4).
## Chain 4:
## Chain 4: Gradient evaluation took 4.3e-05 seconds
## Chain 4: 1000 transitions using 10 leapfrog steps per transition would take 0.43 seconds.
## Chain 4: Adjust your expectations accordingly!
## Chain 4:
## Chain 4:
## Chain 4: Iteration: 1 / 6000 [ 0%] (Warmup)
## Chain 4: Iteration: 600 / 6000 [ 10%] (Warmup)
## Chain 4: Iteration: 1200 / 6000 [ 20%] (Warmup)
## Chain 4: Iteration: 1800 / 6000 [ 30%] (Warmup)
## Chain 4: Iteration: 2001 / 6000 [ 33%] (Sampling)
## Chain 4: Iteration: 2600 / 6000 [ 43%] (Sampling)
## Chain 4: Iteration: 3200 / 6000 [ 53%] (Sampling)
## Chain 4: Iteration: 3800 / 6000 [ 63%] (Sampling)
## Chain 4: Iteration: 4400 / 6000 [ 73%] (Sampling)
## Chain 4: Iteration: 5000 / 6000 [ 83%] (Sampling)
## Chain 4: Iteration: 5600 / 6000 [ 93%] (Sampling)
## Chain 4: Iteration: 6000 / 6000 [100%] (Sampling)
## Chain 4:
## Chain 4: Elapsed Time: 2.05636 seconds (Warm-up)
## Chain 4: 20.156 seconds (Sampling)
## Chain 4: 22.2123 seconds (Total)
## Chain 4:## Warning: There were 10422 divergent transitions after warmup. Increasing adapt_delta above 0.8 may help. See
## http://mc-stan.org/misc/warnings.html#divergent-transitions-after-warmup## Warning: Examine the pairs() plot to diagnose sampling problemsprint(fit_3)## Inference for Stan model: bayes_model_3.
## 4 chains, each with iter=6000; warmup=2000; thin=1;
## post-warmup draws per chain=4000, total post-warmup draws=16000.
##
## mean se_mean sd 2.5% 25% 50% 75%
## intercept_mu 0.58 0.01 0.27 0.05 0.37 0.61 0.82
## intercept_sd 1.35 0.03 0.62 0.25 0.89 1.32 1.75
## intercept[1] 1.65 0.02 1.02 -0.06 0.92 1.55 2.28
## intercept[2] 1.92 0.03 1.23 0.00 1.03 1.76 2.67
## intercept[3] 0.40 0.02 1.10 -1.89 -0.23 0.46 1.07
## intercept[4] -0.62 0.04 1.40 -3.73 -1.44 -0.42 0.39
## intercept[5] 0.45 0.02 0.99 -1.61 -0.15 0.49 1.07
## intercept[6] 1.08 0.02 1.12 -1.00 0.35 0.98 1.75
## intercept[7] 1.59 0.03 1.01 -0.11 0.89 1.49 2.21
## intercept[8] 1.08 0.02 1.01 -0.77 0.43 0.98 1.66
## intercept[9] 1.79 0.03 1.19 -0.12 0.91 1.65 2.51
## intercept[10] -0.35 0.03 1.39 -3.63 -1.12 -0.15 0.60
## intercept[11] -0.12 0.02 1.04 -2.42 -0.75 0.00 0.60
## intercept[12] 1.14 0.02 0.95 -0.55 0.51 1.05 1.70
## intercept[13] 0.07 0.03 1.43 -3.25 -0.66 0.27 0.95
## intercept[14] -0.06 0.02 1.09 -2.50 -0.70 0.06 0.67
## intercept[15] -0.58 0.04 1.45 -4.03 -1.39 -0.35 0.46
## intercept[16] 0.21 0.02 1.04 -1.99 -0.41 0.29 0.87
## intercept[17] 1.61 0.03 1.15 -0.32 0.80 1.48 2.30
## intercept[18] 0.82 0.02 1.20 -1.62 0.13 0.78 1.50
## intercept[19] -0.64 0.04 1.42 -3.94 -1.46 -0.42 0.41
## intercept[20] 0.30 0.02 1.17 -2.22 -0.36 0.40 1.01
## intercept[21] 0.17 0.02 1.44 -3.05 -0.58 0.32 1.02
## intercept[22] 1.01 0.02 0.92 -0.73 0.42 0.96 1.56
## intercept[23] 1.73 0.03 1.15 -0.17 0.90 1.60 2.45
## intercept[24] 0.76 0.02 1.09 -1.41 0.10 0.73 1.41
## intercept[25] 0.96 0.02 1.12 -1.18 0.26 0.89 1.62
## intercept[26] -0.38 0.03 1.37 -3.64 -1.13 -0.18 0.56
## beta_mu -0.74 0.01 0.27 -1.31 -0.90 -0.73 -0.56
## beta_sd 0.93 0.01 0.32 0.38 0.72 0.91 1.12
## beta[1] -0.71 0.01 0.50 -1.76 -1.01 -0.68 -0.37
## beta[2] -0.37 0.01 0.59 -1.57 -0.74 -0.36 0.01
## beta[3] -1.38 0.02 0.71 -2.97 -1.78 -1.31 -0.89
## beta[4] -1.63 0.02 0.90 -3.65 -2.14 -1.54 -1.00
## beta[5] -1.02 0.01 0.58 -2.27 -1.37 -0.99 -0.63
## beta[6] 0.33 0.01 0.66 -0.81 -0.13 0.28 0.71
## beta[7] -0.23 0.01 0.53 -1.29 -0.58 -0.24 0.10
## beta[8] -0.05 0.01 0.49 -1.01 -0.37 -0.05 0.27
## beta[9] -0.61 0.01 0.54 -1.76 -0.94 -0.59 -0.24
## beta[10] -1.51 0.02 0.88 -3.59 -1.99 -1.40 -0.92
## beta[11] -0.95 0.01 0.55 -2.05 -1.31 -0.94 -0.59
## beta[12] -1.42 0.02 0.65 -2.88 -1.80 -1.36 -0.97
## beta[13] -1.26 0.02 0.84 -3.24 -1.73 -1.15 -0.69
## beta[14] -0.65 0.01 0.51 -1.66 -0.98 -0.64 -0.32
## beta[15] -1.60 0.02 0.85 -3.52 -2.09 -1.52 -1.01
## beta[16] -0.55 0.01 0.49 -1.52 -0.87 -0.55 -0.24
## beta[17] -0.12 0.01 0.51 -1.16 -0.45 -0.12 0.21
## beta[18] -1.10 0.01 0.73 -2.75 -1.53 -1.05 -0.61
## beta[19] -1.69 0.02 0.86 -3.61 -2.18 -1.61 -1.09
## beta[20] 0.07 0.01 0.49 -0.84 -0.26 0.05 0.37
## beta[21] -1.24 0.02 0.80 -3.11 -1.69 -1.13 -0.71
## beta[22] -1.22 0.01 0.54 -2.42 -1.55 -1.19 -0.85
## beta[23] -0.01 0.01 0.51 -1.01 -0.35 0.00 0.32
## beta[24] 0.65 0.01 0.64 -0.52 0.22 0.60 1.02
## beta[25] -0.34 0.01 0.59 -1.52 -0.72 -0.34 0.05
## beta[26] -1.49 0.02 0.87 -3.47 -1.96 -1.39 -0.89
## lp__ -132.11 0.71 13.23 -155.09 -140.49 -133.27 -125.18
## 97.5% n_eff Rhat
## intercept_mu 0.98 1552 1
## intercept_sd 2.69 541 1
## intercept[1] 3.85 1729 1
## intercept[2] 4.73 1362 1
## intercept[3] 2.52 3066 1
## intercept[4] 1.60 1540 1
## intercept[5] 2.40 2635 1
## intercept[6] 3.53 2431 1
## intercept[7] 3.82 1616 1
## intercept[8] 3.27 2958 1
## intercept[9] 4.53 1355 1
## intercept[10] 1.86 1983 1
## intercept[11] 1.65 1800 1
## intercept[12] 3.20 1986 1
## intercept[13] 2.54 2303 1
## intercept[14] 1.86 2201 1
## intercept[15] 1.59 1478 1
## intercept[16] 2.18 2575 1
## intercept[17] 4.19 1643 1
## intercept[18] 3.39 2712 1
## intercept[19] 1.56 1464 1
## intercept[20] 2.51 2690 1
## intercept[21] 2.78 3428 1
## intercept[22] 2.96 2273 1
## intercept[23] 4.23 1554 1
## intercept[24] 3.07 2371 1
## intercept[25] 3.32 3240 1
## intercept[26] 1.83 1559 1
## beta_mu -0.26 1408 1
## beta_sd 1.64 1252 1
## beta[1] 0.20 2092 1
## beta[2] 0.81 2150 1
## beta[3] -0.20 2208 1
## beta[4] -0.15 1925 1
## beta[5] 0.05 2739 1
## beta[6] 1.78 2232 1
## beta[7] 0.80 1536 1
## beta[8] 0.91 3057 1
## beta[9] 0.39 1718 1
## beta[10] -0.12 2157 1
## beta[11] 0.11 2255 1
## beta[12] -0.28 1467 1
## beta[13] 0.13 2355 1
## beta[14] 0.38 2498 1
## beta[15] -0.19 2209 1
## beta[16] 0.42 2352 1
## beta[17] 0.87 2153 1
## beta[18] 0.18 2840 1
## beta[19] -0.25 1630 1
## beta[20] 1.10 2810 1
## beta[21] 0.03 2739 1
## beta[22] -0.29 1942 1
## beta[23] 1.00 1950 1
## beta[24] 2.05 1988 1
## beta[25] 0.84 3469 1
## beta[26] -0.08 2045 1
## lp__ -101.60 346 1
##
## Samples were drawn using NUTS(diag_e) at Fri Jan 4 16:50:44 2019.
## For each parameter, n_eff is a crude measure of effective sample size,
## and Rhat is the potential scale reduction factor on split chains (at
## convergence, Rhat=1).posterior_3 <- as.array(fit_3)
mcmc_areas(posterior_3, regex_pars = c("beta"), prob = 0.8, prob_outer = 0.95, point_est = "mean")
Диагностика
posterior_3b <- ggs(fit_3)
ggmcmc(D = posterior_3b, file = NULL, plot = 'ggs_traceplot') ## Plotting traceplots
## Time taken to generate the report: 26 seconds.ggmcmc(D = posterior_3b, file = NULL, plot = 'ggs_compare_partial') ## Plotting comparison of partial and full chain
## Time taken to generate the report: 40 seconds.ggmcmc(D = posterior_3b, file = NULL, plot = 'ggs_autocorrelation') ## Plotting autocorrelation plots
## Time taken to generate the report: 31 seconds.